===================================
Incremental View Synthesis -- Supplementary Material
===================================

> Anonymous NeurIPS submission -- Paper ID 936


Content 
---------

1. Supplementary document (./Incremental Scene Synthesis - Sup Mat.pdf)

    This PDF file complements the main paper. It introduces further pipeline details for reproducibility, as well as provides various additional qualitative results and quantitative comparisons on 2D and 3D datasets.

2. Video of qualitative results (./Incremental Scene Synthesis - Qualitative Results.avi)

    This video file presents several qualitative results. Our pipeline is applied to the incremental generation of complete environments, given a sequence of unlocalized partial views. Experiments are done according to the setups described in the main paper appendix, walking different agents (with various behaviors and view fields) on top of images from the CelebA [1] and HoME-2D [2] datasets.

	
Codec
-------

The video is encoded with H.264, which can be downloaded here:
https://www.codecguide.com/download_kl.htm


References
-------------
[1] Z. Liu, P. Luo, X. Wang, and X. Tang. Deep learning face attributes in the wild.   In Proceedings of International Conference on Computer Vision (ICCV), December 2015.
[2] S. Brodeur,  E. Perez, A. Anand, F. Golemo, L. Celotti,F. Strub, J. Rouat, H. Larochelle, and A. Courville. HoME: A household multimodal environment. arXiv preprintarXiv:1711.11017, 2017.
